The present invention relates to manufacturing microelectronic, microoptoelectronic, or micromechanical devices.
Microelectronic devices (also called integrated electronic circuits, or ICs) are the base of the integrated electronics industry. Microoptoelectronic devices comprise, for example, new generations of infrared radiation (IR) sensors which, unlike traditional ones, do not require cryogenic temperatures for their operation. These IR sensors are formed of an array of semiconductor material deposits, for example silicon, arranged in an evacuated chamber. Micromechanical devices (better known in the field under the definition “micromachines” or referred herein as MMs) are being developed for applications such as miniaturized sensors or actuators. Typical examples of micromachines are microaccelerometers, which are used as sensors to activate automobile airbags; micromotors, having gears and sprocket wheels of the size of a few microns (um); or optic switches, wherein a mirror surface with a size of the order of a few tens microns can be moved between two different positions, directing a light beam along two different directions, one corresponding to the “on” condition and the other to the “off” condition of an optical circuit. In the following description, these devices will also all be referred to within the general definition of solid-state devices.
ICs are manufactured by depositing layers of material with different electric (or magnetic) functionalities on a planar then selectively removing them to create the device. The same techniques of depositions and selective removal create microoptoelectronic or micromechanical devices as well. These devices are generally contained in housings formed, in their turn, with the same techniques. The support most commonly used in these productions is a silicon “slice” (usually referred to as a “wafer”), about 1 mm thick and with a diameter up to 30 cm. On each of these wafers a very high number of devices may be constructed. At the end of the manufacturing process, individual devices, in the case of micromachines, or parts, in the case of IR sensors, are separated from the slices using mechanical or laser means.
The deposition steps are carried out with such techniques as, for example, chemical deposition from vapor state, (“Chemical Vapor Deposition” or “CVD”) and physical deposition from vapor state (“Physical Vapor Deposition” or “PVD”). The latter is commonly known in the art as “sputtering.” Generally, selective removals are carried out through chemical or physical attacks using proper masking techniques. Such techniques are well-known in the field and will not be discussed here except as they relate to specific embodiments of the invention.
The integrated circuits and the micromachines are then encapsulated in polymeric, metallic or ceramic materials, essentially for mechanical protection, before being put to final use (within a computer, an automobile, etc.). In contract, IR radiation sensors are generally encapsulated in a chamber, facing one wall thereof, transparent to the IR radiation and known as a “window.”
In certain integrated circuits it is important to be able to control the gas diffusion in solid state devices. For example, in the case of ferroelectric memories, hydrogen diffuses through device layers and can reach the ferroelectric material, which is generally a ceramic oxide, such as lead titanate-zirconate, strontium-bismuth tantalate or titanate, or bismuth-lanthanum titanate. When the hydrogen reaches the ferroelectric material, it can alter its correct functioning.
Still more important is gas control and elimination in IR sensors and in micromachines. In the case of IR sensors, the gases which may be present in the chamber can either absorb part of the radiation or transport heat by convection from the window to the array of silicon deposits, altering the correct measurement. In the case of micromachines, the mechanical friction between gas molecules and the moving part, due to the very small size of the latter, can lead to detectable deviations from the device's ideal operation. Moreover, polar molecules such as water can cause adhesion between the moving part and other parts, such as the support, thus causing the device's failure. In the IR sensors with arrays of silicon deposits or in the micromachines, it is therefore fundamental to ensure the housing remains in vacuum for the whole device life.
In order to minimize the gas amount in these devices, their production is usually conducted in vacuum chambers and resorting to pumping steps before the packaging. However, the problem is not completely solved by pumping because the same materials which form the devices can release gases, or gases can permeate from outside during the device life.
To remove the gases entering in solid state devices during their life the use of materials that can sorb these destructive gases may be helpful. These absorptive materials are commonly referred to as “getters,” and are generally metals such as zirconium, titanium, vanadium, niobium or tantalum, or alloys thereof combined with other transition elements, rare earths or aluminum. Such materials have a strong chemical affinity towards gases such as hydrogen, oxygen, water, carbon oxides and in some cases nitrogen. The absorptive materials also include the drier materials, which are specifically used for moisture absorption, which usually include the oxides of alkali or alkaline-earth metals. The use of materials for absorbing gases, particularly hydrogen, in ICs, is described for instance in U.S. Pat. No. 5,760,433 by Ramer et al. Ramer teaches that the chemically reactive getter material is formed as part of the process of fabricating the integrated circuit. The use of getters in IR sensors is described in U.S. Pat. No. 5,921,461 by Kennedy et al. Kennedy teaches that a getter is deposited onto preselected regions of the interior of the package. Finally, the use of gas absorbing materials in micromachines is described in the article “Vacuum packaging for microsensors by glass-silicon anodic bonding” by H. Henmi et al., published in the technical journal Sensors and Actuators A, vol. 43 (1994), at pages 243–248.
The above references teach that localized deposits of gas absorbing materials can be obtained by CVD or sputtering during solid-state device production steps. However, this procedure can be costly and time consuming if done during the solid-state manufacturing CVD or sputtering process. This is because gas absorbing material deposition during device production implies the step involved in localized deposition of the gas absorbing or getter material. This is generally carried out through the steps of resin deposition, resin local sensitization through exposure to radiation (generally UV), selective removal of the photosensitized resin, gas absorbing material deposition and subsequent removal of the resin and of the absorbing material thereon deposed, leaving the gas absorbing material deposit in the area in which the photosensitized resin had been removed. Moreover, depositing the gas absorbing material in the production line is disadvantageous because there are an increased number of steps required in the manufacturing process. Increasing deposits, in turn, requires that more materials be used, which also significantly increases the risk of “cross-contamination” among the different chambers in which the different steps are carried out. Also, there is a possible increase of waste products because of contamination.